'labels' as a DataFrame column name in version 0.14.1

[FragLegs]

I'm not sure if this qualifies as a bug, but it is a change that occurred between 0.13.1 and 0.14.1. In 0.14.1, if I have a DataFrame with a column named 'labels', I get the following error when trying to access it as an attribute of the DataFrame:

>>> df = pandas.DataFrame(columns=['labels'])
>>> df['labels']
Series([], name: labels, dtype: object)
>>> df.labels
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/<virtual_env>/local/lib/python2.7/site-packages/pandas/core/generic.py", line 1836, in __getattr__
    return object.__getattribute__(self, name)
AttributeError: 'DataFrame' object has no attribute 'labels'

In 0.13.1, this code did not throw an error. This is with Python 2.7.6.

[jreback]

this section should be updated a bit: http://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access

Their are a couple of 'reserved' names that are checked before the actual column names (this is done for efficiency). This could technically be avoided, but would require a bit of code to do that.

e.g.

index,major_axis,minor_axis,items,labels iow the names of the axes.

So these work with df['index'] but not attribute access.

care to do a pull-request to update?

[jreback]

closed by #8100